It is known that combinatorial computation and the like operations are performed by a microcomputer in a combinatorial weighing (or counting) apparatus which computes combinations of sets of objects being weighed which vary in weight, retrieves the combination which satisfies a set weight value and discharges it or which computes combinations of sets of objects being counted which vary in number, retrieves the combination which satisfies a set number and discharges it. When objects discharged by such system are later examined by separate weighing (or counting), it is sometimes found that they have been discharged with a value that considerably deviates from the set value. However, it is usual with this type of apparatus that it is utterly impossible to check the computations performed by the microcomputer. Therefore, even if such a trouble occurs, there is no ascertaining whether the trouble is due to errors on the part of the microcomputer or the weighing machines. Further, even if the discharged objects are weighed (or counted) again for check up and found to satisfy the set value, it has been still unknown whether the value is actually the nearest to the set value. More particularly, if a quantity of objects whose weight (or number) is exactly equal to the set value are discharged, there would be no problem, of course, but if a quantity of objects whose weight (or number) is more or less deviated from the set value are discharged, there is no knowing whether the combination of objects whose value is actually the nearest to the set value has been discharged or a combination of objects whose value is, e.g., the second nearest to the set value has been discharged owing to a mulfunction of the microcomputer.